Experimental Petrology of Olivine-phyric Shergottite Nwa 1068: Toward

Introduction: The olivine-phyric basaltic shergottites (OPSs) have recently been recognized as a significant sub-group of the Martian meteorites. These rocks have many characteristics suggestive of a primitive nature such as high bulk and olivine-core Mg#s [1]. [2,3] conducted crystallization experiments using the composition of Yamato 980459 (Y98), an OPS. Results of these experiments show that the olivine cores are in equilibrium with the bulk rock, indicating that the bulk-rock composition is the same as the parent melt. Multiple saturation of the melt with olivine and a low-calcium pyroxene occurs at 11.5 to 12 kbars and 1450 ± 10 °C indicating that the melt separated from the mantle at a depth of ~100 km. Experiments similar to [2,3] on the Gusev Crater basalt composition [4] found multiple saturation at a pressure of 11kbar -nearly identical to the pressure for Y98, implying a similar lithospheric thickness under Gusev Crater. The bulk composition of NWA 1068, another OPS, may also represent something close to a primary mantle-derived magma. Although some evidence suggests modification of the bulk composition from the original parent magma by accumulation of olivine [5], its texture is one of rapid crystal growth just prior to emplacement. NWA1068 is a significant specimen as it is the only olivinephyric shergottite that is not LREE depleted, having an REE pattern [5], Sm-Nd and Rb-Sr isotopic systematics [6], and fO2 [7] similar to Shergotty. Intriguingly, calculations using the MELTS program [8] predict multiple saturation for the NWA bulk composition at a similar pressure to the Y98 and Gusev compositions. For these reasons we have undertaken experiments at Martian upper mantle conditions to explore its phase chemistry. Experimental Methods: C r y s t a l l i z a t i o n experiments were conducted using a piston cylinder apparatus on a synthetic NWA1068 composition at Martian mantle pressures and temperatures. The starting materials were made following the method of [3]. The starting composition for the experiments (Table 1) is based on the analysis of [5]. This analysis was by ICP-MS with no SiO2 analysis, so SiO2 was determined by subtraction. To correct for the effect of terrestrial weathering, 1% of the CaO as carbonate was assumed to be from terrestrial weathering and subtracted out and a nominal amount of P2O5 was assumed. The resulting adjusted starting composition is given in Table 1 along with the composition calculated from actual weights of oxides and carbonates and electron microprobe analysis of the run product from an experiment that produced only glass. Oxygen fugacity in the piston-cylinder experiments was controlled using the method of [9], whereby starting materials with preset Fe2O3 [10, 11] are run in graphite capsules. The resulting fO2 is IW +2.5 ± 0.2 log units. The run procedures are the same as [3]. Pressure was calibrated against the melting point of diopside [12]. The pressure offset is 0.8±0.1 kbar below gauge. One-atmosphere experiments were conducted following the procedure of [3]. All run products were analyzed using the Cameca SX-100 electron microprobe at NASA/JSC.